vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix: examples/swiftui llama_model -> llama_vocab

Browse source
This commit is contained in:
Maksym Matviievskyi 2025-02-03 20:10:26 +00:00
parent 5598f475be
commit 5cb76bc193
2 changed files with 7 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

6
examples/batched.swift/Sources/main.swift
View file

@ -207,7 +207,7 @@ private func tokenize(text: String, add_bos: Bool) -> [llama_token] {
let utf8Count = text.utf8.count let utf8Count = text.utf8.count
let n_tokens = utf8Count + (add_bos ? 1 : 0) let n_tokens = utf8Count + (add_bos ? 1 : 0)
let tokens = UnsafeMutablePointer<llama_token>.allocate(capacity: n_tokens) let tokens = UnsafeMutablePointer<llama_token>.allocate(capacity: n_tokens)
let tokenCount = llama_tokenize(model, text, Int32(utf8Count), tokens, Int32(n_tokens), add_bos, /*special tokens*/ false) let tokenCount = llama_tokenize(llama_model_get_vocab, text, Int32(utf8Count), tokens, Int32(n_tokens), add_bos, /*special tokens*/ false)
var swiftTokens: [llama_token] = [] var swiftTokens: [llama_token] = []
for i in 0 ..< tokenCount { for i in 0 ..< tokenCount {
swiftTokens.append(tokens[Int(i)]) swiftTokens.append(tokens[Int(i)])
@ -218,12 +218,12 @@ private func tokenize(text: String, add_bos: Bool) -> [llama_token] {
private func token_to_piece(token: llama_token, buffer: inout [CChar]) -> String? { private func token_to_piece(token: llama_token, buffer: inout [CChar]) -> String? {
var result = [CChar](repeating: 0, count: 8) var result = [CChar](repeating: 0, count: 8)
let nTokens = llama_token_to_piece(model, token, &result, Int32(result.count), 0, false) let nTokens = llama_token_to_piece(llama_model_get_vocab, token, &result, Int32(result.count), 0, false)
if nTokens < 0 { if nTokens < 0 {
let actualTokensCount = -Int(nTokens) let actualTokensCount = -Int(nTokens)
result = .init(repeating: 0, count: actualTokensCount) result = .init(repeating: 0, count: actualTokensCount)
let check = llama_token_to_piece( let check = llama_token_to_piece(
model, llama_model_get_vocab,
token, token,
&result, &result,
Int32(result.count), Int32(result.count),

8
examples/llama.swiftui/llama.cpp.swift/LibLlama.swift
View file

@ -151,7 +151,7 @@ actor LlamaContext {
new_token_id = llama_sampler_sample(sampling, context, batch.n_tokens - 1) new_token_id = llama_sampler_sample(sampling, context, batch.n_tokens - 1)
if llama_vocab_is_eog(model, new_token_id) || n_cur == n_len { if llama_vocab_is_eog(llama_model_get_vocab, new_token_id) || n_cur == n_len {
print("\n") print("\n")
is_done = true is_done = true
let new_token_str = String(cString: temporary_invalid_cchars + [0]) let new_token_str = String(cString: temporary_invalid_cchars + [0])
@ -297,7 +297,7 @@ actor LlamaContext {
let utf8Count = text.utf8.count let utf8Count = text.utf8.count
let n_tokens = utf8Count + (add_bos ? 1 : 0) + 1 let n_tokens = utf8Count + (add_bos ? 1 : 0) + 1
let tokens = UnsafeMutablePointer<llama_token>.allocate(capacity: n_tokens) let tokens = UnsafeMutablePointer<llama_token>.allocate(capacity: n_tokens)
let tokenCount = llama_tokenize(model, text, Int32(utf8Count), tokens, Int32(n_tokens), add_bos, false) let tokenCount = llama_tokenize(llama_model_get_vocab, text, Int32(utf8Count), tokens, Int32(n_tokens), add_bos, false)
var swiftTokens: [llama_token] = [] var swiftTokens: [llama_token] = []
for i in 0..<tokenCount { for i in 0..<tokenCount {
@ -316,7 +316,7 @@ actor LlamaContext {
defer { defer {
result.deallocate() result.deallocate()
} }
let nTokens = llama_token_to_piece(model, token, result, 8, 0, false) let nTokens = llama_token_to_piece(llama_model_get_vocab, token, result, 8, 0, false)
if nTokens < 0 { if nTokens < 0 {
let newResult = UnsafeMutablePointer<Int8>.allocate(capacity: Int(-nTokens)) let newResult = UnsafeMutablePointer<Int8>.allocate(capacity: Int(-nTokens))
@ -324,7 +324,7 @@ actor LlamaContext {
defer { defer {
newResult.deallocate() newResult.deallocate()
} }
let nNewTokens = llama_token_to_piece(model, token, newResult, -nTokens, 0, false) let nNewTokens = llama_token_to_piece(llama_model_get_vocab, token, newResult, -nTokens, 0, false)
let bufferPointer = UnsafeBufferPointer(start: newResult, count: Int(nNewTokens)) let bufferPointer = UnsafeBufferPointer(start: newResult, count: Int(nNewTokens))
return Array(bufferPointer) return Array(bufferPointer)
} else { } else {